Trouble understanding indexing while looping through 2d lists

Question

Maybe I am just not understanding list indexing correctly, but when I run the following code:

myList= [['nan'] *4]*4  

for i in range(4):
  for j in range(4):
    if [i,j] == [2,1]: 
      myList[i][j]=1
    print('['+str(i)+','+str(j)+']  -->   '+str(myList[i][j]))

I believe it should output

[0][0]  -->   nan
[0][1]  -->   nan
[0][2]  -->   nan
[0][3]  -->   nan
[1][0]  -->   nan
[1][1]  -->   nan
[1][2]  -->   nan
[1][3]  -->   nan
[2][0]  -->   nan
[2][1]  -->   1
[2][2]  -->   nan
[2][3]  -->   nan
[3][0]  -->   nan
[3][1]  -->   nan
[3][2]  -->   nan
[3][3]  -->   nan

However, it actually outputs

[0][0]  -->   nan
[0][1]  -->   nan
[0][2]  -->   nan
[0][3]  -->   nan
[1][0]  -->   nan
[1][1]  -->   nan
[1][2]  -->   nan
[1][3]  -->   nan
[2][0]  -->   nan
[2][1]  -->   1
[2][2]  -->   nan
[2][3]  -->   nan
[3][0]  -->   nan
[3][1]  -->   1
[3][2]  -->   nan
[3][3]  -->   nan

What's more, if I do

myList= [['nan'] *4]*4 

for i in range(4):
  for j in range(4):
    if i==1 and j==2: 
      myList[i][j]=1
    print('['+str(i)+']['+str(j)+']  -->   '+str(myList[i][j]))

I get

[0][0]  -->   nan
[0][1]  -->   nan
[0][2]  -->   nan
[0][3]  -->   nan
[1][0]  -->   nan
[1][1]  -->   nan
[1][2]  -->   1
[1][3]  -->   nan
[2][0]  -->   nan
[2][1]  -->   nan
[2][2]  -->   1
[2][3]  -->   nan
[3][0]  -->   nan
[3][1]  -->   nan
[3][2]  -->   1
[3][3]  -->   nan

Why is it in the first case changing the value at [3][1], and in the later case at [1][2] and [3][2]? The if statement should never be reached for any index except [2][1] in either case! Also, why is are the two if statements not equivalent?

When you do `myList= [['nan'] *4]*4` you create the inner list with `['nan'] *4` and then you add this one list 4 times to the outer list. Since all your inner lists are the same list changing a value in one will of course reflect on the others. — Matthias, Aug 03 '21 at 19:13

Nothing special · Accepted Answer · 2021-08-03T19:30:38.523

Actually you are correct about indexing, the trouble is with your very first statement.

myList = [['nan'] *4]*4

Each of your inner list ['nan', 'nan', 'nan', 'nan'] is referring to the same object. When you are changing a value in one of the inner lists, this changes your object. As the object changes, all the other inner lists change.

The solution to this problem is simple. Don't use *. This is a pythonic way of solving the problem:

myList = [['nan' for i in range(4)] for i in range(4)]

Un-pythonic way of doing this:

myList = []
for i in range(4):
    myList.append([])
    for j in range(4):
        myList[i].append('nan')

I am sorry for the inconvenience. I've edited the answer. – Nothing special Aug 03 '21 at 19:23 — Nothing special, Aug 03 '21 at 19:23

score 1 · Answer 2 · answered Aug 03 '21 at 19:14

Two different issues. The if statements are different because you reversed the order.

if i==1 and j==2: is if [i,j] == [1,2]:

and if [i,j] == [2,1]: is if i==2 and j==1:

As for the other issue.. you're probably not familiar with mutability. The three lists you have inside of the top level list is actually 3 references to the same object.

This question shows how to print a memory address: print memory address of Python variable

Try it on myList, like print(hex(id(myList[0])), 'next', hex(id(myList[1])), 'next', hex(id(myList[1])))

They all have the same address. So when you modify one, it modifies all three, you just don't see the first because you've already gone past it.

Thank you that makes sense! – user1093541 Aug 03 '21 at 20:19 — user1093541, Aug 03 '21 at 20:19

Trouble understanding indexing while looping through 2d lists

2 Answers2